Predicting Failure of Conservative Airway Management in Infants with Robin Sequence: The EARN Factors

Abstract

Objective

Evaluate infants with Robin Sequence (RS) who were successfully treated with conservative airway measures alone vs. those who failed and eventually underwent surgical airway intervention after a protracted course of conservative management.

Design

Retrospective review of prospectively gathered database.

Setting

Large tertiary care institution.

Patients

Infants diagnosed with RS (n = 122) who underwent primary airway management at a single institution from 1994–2020.

Main Outcome Measure

Patient demographics, nutritional and respiratory status, laboratory values, and polysomnographic results were compared between patients who were discharged after successful conservative airway management (Group 1, n = 61) and patients that underwent surgical airway intervention after failing a prolonged course of conservative management (Group 2, n = 61). Receiver operating characteristic (ROC) curve analysis was done to assess continuous variables that may predict failure of conservative airway management.

Results

122 infants with RS were investigated. While several variables were significantly different between groups, the following polysomnographic EARN factors, with cut points, were identified as most predictive of failed conservative airway management: ETCO2 (max) > 49 mmHg, AHI > 16.9 events/hour, OAHI REM >25.9 events/hour, OAHI Non-REM > 23.6 events/hour.

Conclusions

We identified factors in infants with RS that were associated with severe UAO that failed to improve despite weeks of conservative airway management. Our results may expedite earlier definitive treatment of these critical patients and reduce risks for known complications of prolonged UAO.

Keywords

robin sequence upper airway obstruction TLA MDO tracheostomy polysomnogram

Introduction

Robin sequence (RS), the clinical triad of micrognathia, glossoptosis, and upper airway obstruction (UAO), affects approximately 1 in 8500 births.^1,2 Infants may present with wide phenotypic variability, ranging from infrequent respiratory and/or feeding difficulties to severe UAO and failure to thrive.³ Treatment options range from conservative modalities, namely prone positioning, oropharyngeal hygiene, nutritional support, and nasopharyngeal intubation,^4–6 to operative interventions, such as subperiosteal release of floor of mouth,⁷ tongue-lip adhesion,^8–14 tracheostomy,^15,16 and mandibular distraction osteogenesis.^17–24

Pediatric sleep apnea has both short-term and long-term deleterious effects on multiple organ systems and the importance of early diagnosis and treatment of UOA in infants with RS has been increasingly recognized. Several studies have documented neurobehavioral and neurocognitive deficits associated with pediatric obstructive sleep apnea (OSA),^25–28 which some argue are irreversible if OSA is left untreated.²⁹ Hypertension, increased heart rate, impaired left ventricular function, and changes in right ventricular dimensions have also been reported in children with long-standing OSA.^30–32

At our institution, infants with RS are managed by a multidisciplinary team that provides a holistic and deliberate approach to the management of these complex patients. Except in cases of immediate, emergent airway compromise, every effort is made to treat UAO using conservative measures alone. Nevertheless, our group and others have demonstrated that some patients who may initially improve with conservative interventions, ultimately fail this approach and require a surgical airway procedure.^33,34 While our institution previously published clinical factors associated with successful conservative airway management in infants with RS,³⁵ a direct comparison with patients who failed conservative airway treatment was not performed. To establish an evidence-based set of guidelines, we therefore designed a study to review airway management in patients with RS at our tertiary care center and identify clinical factors that may predict failure of conservative airway management in these complex patients and avoid an unnecessary, prolonged preoperative course.

Methods

A retrospective review of a prospectively gathered database was undertaken for patients with RS who underwent primary airway management at a large tertiary care hospital from 1994–2020. After obtaining IRB approval, the study cohort was identified by performing a search within the billing departments for the Division of Plastic and Reconstructive Surgery, as well as Otolaryngology and Neonatology, using International Classification of Diseases, version 9 (ICD-9) codes for anomaly of jaw size and related diagnoses (524.00, 524.06, 524.10) as well as Current Procedure Terminology (CPT) codes for tongue-lip adhesion (TLA), mandibular distraction osteogenesis (MDO), and tracheostomy (41510, 20690/20692, and 31600/31601/31603/31605, respectively). In addition, a review of an IRB approved, prospectively gathered database for patients with RS maintained by the Division of Plastic and Reconstructive Surgery was performed to cross-reference with billing data.

Inclusion criteria consisted of patients with RS who were managed at our institution. When available, polysomnogram (PSG) data was included if it was completed within 90 days before admission, during the hospital stay or before any surgical treatment was performed. In patients who were discharged after successful conservative management, PSG was included if performed within 90 days of discharge. Patients with micrognathia and non-obstructive respiratory failure (eg, congestive heart failure, renal failure), as well as those who received definitive airway management at an outside hospital, were excluded from the study.

Information recorded from all cases included demographic characteristics, preexisting medical conditions and/or associated syndromic diagnosis, consecutive weight measurements during admission, PSG (when available), surgical treatment and discharge. In addition, hospital factors such as type of nutritional gavage feeding, type of respiratory support, and desaturation events were recorded.

Maximum titratable bicarbonate (maxHCO₃) and maximum carbon dioxide (maxCO₂) were drawn via capillary blood gas (CBG). PSG data were obtained as previously described by our group³⁶ and the following variables were collected: Total apnea-hypopnea index (AHI), obstructive apnea-hypopnea index (OAHI), central apnea index (CAI), lowest oxygen saturation (%), sleep time with O₂ saturation < 90%, mean ETCO₂ and maximum ETCO₂.

Patients were divided into two groups, based on definitive airway management. Group 1 patients were discharged from the hospital after successful treatment using only conservative measures that included lateral/prone positioning, oropharyngeal hygienic support, enteric nutritional feeding, and respiratory support, including nasopharyngeal tube insertion by otolaryngology. Patients in Group 2 failed a prolonged course of conservative treatment and underwent surgical airway procedures, including tracheostomy, TLA, and MDO.

Multidisciplinary Team Protocol

All infants with suspected RS at CNH are managed by a multidisciplinary team protocol. For in-patients, consultation from key team members is obtained with 24–48 h of admission, including plastic surgery, otolaryngology, genetics, pulmonary medicine, general surgery, as well as speech/feeding therapy. Bedside flexible endoscopy is performed to assess the airway. If patients have not shown significant improvement with conservative airway management after several days, they undergo direct laryngoscopy/bronchoscopy (DLB) and drug induced sleep endoscopy (DISE), as well as PSG. Conservative measures were attempted for as long as possible, until both the family and multidisciplinary deemed further efforts to be either futile or unsafe. Patients who meet the GILLS criteria^8,37 are offered TLA, in addition to MDO, the latter now being our procedure of choice for most patients. Infants with severe dysmorphia, hypotonia, or multi-level airway obstruction undergo tracheostomy, which is otherwise reserved as a final option.

Statistical Methods

Baseline demographics and hospital factors were compared between the two groups (conservative vs. surgical management). Unpaired t-test and Wilcoxon-Mann-Whitney U tests were used to compare normally distributed and skewed continuous data, respectively. Categorical variables were compared using the Chi-square test and Fisher's exact test (if any of the expected cell sizes was less than 5). Additional analysis was completed in patients with PSG data available. Summary statistics for continuous variables were presented as mean with standard deviation for normally distributed data and median with interquartile range (IQR) for skewed data. Categorical data were summarized as frequencies with percentages. Normality assumptions were checked using a statistical test (eg, Shapiro Wilk test) as well as graphical methods (eg, histogram, q-q plot). A multivariable logistic regression analysis was performed to determine the independent hospital and patient factors associated with failed conservative management. Variables were selected in the multivariable model based on statistical significance (p < 0.05) in the univariate analyses.

To assess the predictive ability of continuous variables found to be significantly different between groups on univariate and multivariable analyses, we performed receiver operating characteristic (ROC) curve analysis. Accuracy was graded based on the area under the curve (AUC) defined as “poor” (AUC 0.60–0.70), “fair” (AUC 0.70–0.80), “good” (AUC 0.80–0.90), and “excellent” (AUC 0.90–1.00).³⁸ The optimal cut-points were calculated using the Youden index (J) method.³⁹ This method calculates the optimal cut-point by maximizing the Youden function which is the difference between true positive rate and false-positive rate over all possible cut-point values. Sensitivity and specificity at the optimum cut point was also calculated. All statistical tests were two-sided and were performed at the 5% level of significance unless otherwise stated. Stata software, version 15.1 MP (Stata Corporation, College Station, Texas, USA) was used for all statistical analyses.

Results

122 infants were included. Syndromic association was noted in roughly half of all patients (50.4%). Demographic results are summarized in Table 1 . Overall hospital course for all patients is summarized in Table 2 . Of the 122 patients, half of the cohort was successfully managed with conservative measures alone (Group 1), while the other half failed nonsurgical intervention and eventually underwent an operative airway procedure (Group 2). Median time to hospital discharge was 12.0 days and 54.0 days (p < 0.001) for Group 1 and Group 2, respectively. In Group 2, median time from PSG to surgery and median time from admission to surgery was 14.0 days (8.0, 22.0) and 23.0 days (13.5, 47.0), respectively. Initial surgical airway management consisted of tracheostomy in 17 patients, TLA in 24 patients, and MDO in 20 patients. Ten of the surgically treated patients required a secondary procedure: 3 infants initially treated with TLA went on to MDO; 3 other TLA patients underwent tracheostomy; and 4 cases of primary MDO had secondary tracheostomy. Baseline patient characteristics are summarized in Table 3 , while analyzed hospital factors are displayed in Table 4 .

Table 1.

Overall Demographics.

Demographic factors (n = 122)	Value
Gestational Weeks, median (IQR)	38.5 (37.0, 40.0)
Birthweight in gram, mean (SD)	2911.8 (745.2)
Gender (Female)	63 (51.6%)
Apgar score, mean (SD)
1-min Apgar score	6.2 (2.4)
5-min Apgar score	7.8 (1.3)
Maternal history
Alcohol	3 (2.5%)
Tobacco	8 (6.7%)
Drugs	9 (7.5%)
Family history
Robin sequence	5 (4.2%)
Stickler syndrome	5 (4.2%)
Cleft Lip	1 (0.8%)
Cleft Palate	10 (8.3%)
Syndromic	61 (50.0%)
Syndromes
Stickler	14 (11.6%)
VCF	1 (0.8%)
Other	48 (39.7%)
Physical anomaly
Cardiac	37 (30.3%)
Pulmonary	28 (23.0%)
CNS	32 (26.2%)
Renal	25 (20.5%)
GERD	68 (55.7%)
Cleft Lip	4 (3.3%)
Cleft Palate	93 (76.2%)
HFM	3 (2.5%)

VCF, velocardiofacial syndrome; CNS, central nervous system; GERD, gastroesophageal reflux; HFM, hemifacial microsomia.

Table 2.

Overall Hospital Course.

Factors (n = 122)	Mean (SD) / median (IQR) / n (%)
Age at visit/ admission(days of life)	8.5 (2.0, 22.0)
Admission weight (kg)	3.3 (1.3)
Time from visit/admission to PSG (days)	12.0 (5.0, 26.0)
Time from admission to discharge / LOS (days)	35.0 (18.0, 65.0)
Time from PSG to discharge (days)	29.0 (17.0, 49.5)
Discharge Weight (kg)	4.0 (1.4)
Surgical Management
Conservative (Group 1)	61 (50.0%)
Surgical (Group 2)	61 (50.0%)

Table 3.

Patient Factors by Type of Airway Management.

Patient factors	Conservative (n = 61)	Surgical (n = 61)	p-value
Patient factors	Mean (SD) / median (IQR) / n (%)		p-value
Gestational Weeks	38.5 (37.0, 40.0)	38.5 (37.0, 40.0)	0.87
Birthweight in gram	2924.9 (748.6)	2898.7 (748.1)	0.85
Gender (Female)	34 (56%)	29 (48%)	0.37
Apgar score
1-min Apgar score	6.6 (2.2)	5.8 (2.5)	0.17
5-min Apgar score	8.3 (0.9)	7.5 (1.5)	0.006
Maternal history
Alcohol	1 (2%)	2 (3%)	0.99
Tobacco	5 (9%)	3 (5%)	0.48
Drugs	4 (7%)	5 (8%)	0.99
Family history
Robin sequence	3 (5%)	2 (3%)	0.68
Stickler syndrome	4 (7%)	1 (2%)	0.20
Cleft Lip	0 (0%)	1 (2%)	0.99
Cleft Palate	5 (8%)	5 (8%)	0.99
Syndromic	28 (46%)	33 (55%)	0.32
Syndromes
Stickler	7 (11%)	7 (12%)	0.99
VCF	0 (0%)	1 (2%)	0.50
Other	19 (31%)	29 (48%)	0.053
Physical anomaly
Cardiac	17 (28%)	20 (33%)	0.55
Pulmonary	8 (13%)	20 (33%)	0.010
CNS	11 (18%)	21 (34%)	0.040
Renal	10 (16%)	15 (25%)	0.26
GERD	28 (46%)	40 (66%)	0.029
Cleft Lip	0 (0%)	4 (7%)	0.12
Cleft Palate	46 (75%)	47 (77%)	0.83
HFM	2 (3%)	1 (2%)	0.99

VCF, velocardiofacial syndrome; CNS, central nervous system anomalies; GERD, gastroesophageal reflux; HFM, hemifacial microsomia.

* P-values were obtained from unpaired t-test/Wilcoxon-Mann-Whitney test for continuous data, and Chi square test/Fisher's exact test for categorical data.

Table 4.

Hospital Factors by Type of Airway Management.

Hospital factors	Conservative (n = 61)	Surgical (n = 61)	p-value
Hospital factors	Mean (SD) / median (IQR) / n (%)		p-value
Age at visit or admission (days of life)	13.0 (2.0, 24.0)	7.0 (2.0, 16.0)	0.23
Length of stay	12 (9.5, 29.0)	54.0 (35.0, 82.0)	<0.001
Admission weight (kg)	3.3 (1.6)	3.2 (1.0)	0.44
Nutritional Status
PO	17 (28%)	3 (5%)	<0.001
NGT	30 (50%)	46 (75%)
PEG	13 (22%)	12 (20%)
Oxygen supplement*
Low flow device	17 (29%)	43 (70%)	<0.001
High flow device	17 (29%)	44 (72%)	<0.001
NPA	4 (7%)	12 (20%)	0.038
Times intubated
No intubation	50 (85%)	37 (61%)	0.011
Once	8 (13%)	19 (31%)
Twice	1 (2%)	5 (8%)
Desaturation event	0.5 (2.8)	0.6 (1.5)	0.94
Capillary Blood Gas
Max Bicarb (mmol/L)	28.0 (8.1)	32.7 (5.5)	0.001
Max CO₂ (mm Hg)	51.4 (14.0)	67.8 (18.2)	<0.001

PO, per os; NGT, nasogastric tube; PEG, percutaneous gastrostomy tube; NPA, nasopharyngeal airway; CO₂, carbon dioxide.

P-values were obtained from unpaired t-test/Wilcoxon-Mann-Whitney test for continuous data, and Chi square test/Fisher's exact test for categorical data.

* These categories were counted as independent variables as some patients had high flow/low flow and NPA during their hospital stay.

To further investigate the independent hospital and patient factors associated with failure of conservative airway management, we performed a multivariable logistic regression analysis on the statistically significant variables identified by univariate analysis above (highlighted in Tables 3 and 4). This multivariate model demonstrated that none of the parameters identified by univariate analysis were statistically correlated with unsuccessful conservative airway measures, with the exception of greater LOS correlating with a surgical airway procedure (adjusted odds ratio 1.03, 95% CI: 1.01–1.06, p = 0.040).

PSG could not be obtained for intubated patients or those on significant O2 support, for obvious reasons (n = 18, 14.8%). In addition, patients who rapidly improved with conservative measures did not undergo PSG (n = 25, 20.5%), at the discretion of the neonatology team. Hence, a total of 79 patients (64.8%) had PSG data available. The median age at PSG was 23.0 days (IQR: 11.0, 49.0 days). As noted above, patients in Group 2 had surgery at a median of 14.0 days (8.0, 22.0) after sleep study was obtained. PSG variables are shown in Table 5 .

Table 5.

PSG Variables by Type of Airway Management.

PSG variables	Conservative (n = 34)	Surgical (n = 45)	p-value
PSG variables	Mean (SD) or Median (IQR)		p-value
Weight (Kg)	3.8 (1.5) (n = 20)	3.2 (0.6) (n = 26)	0.068
AHI or RDI (events/hr)	20.6 (22.8) (n = 29)	59.7 (45.4) (n = 41)	<0.001
OAHI
REM	24.1 (9.8, 51.7) (n = 11)	68.5 (50.8, 133.9) (n = 24)	0.001
Non-REM	8.9 (5.6, 23.1) (n = 11)	40.2 (19.4, 71.2) (n = 24)	0.006
Supine REM	19.4 (8.1, 71.8) (n = 10)	58.0 (32.7, 99.7) (n = 21)	0.12
Non-supine REM	27.2 (9.1, 53.7) (n = 11)	64.0 (36.6, 107.8) (n = 22)	0.018
CAI	0.6 (0.1, 2.6) (n = 24)	0.7 (0.2, 2.6) (n = 35)	0.92
Time O₂ < 90% (%)	1.1 (0.3, 3.6) (n = 23)	10.4 (1.5, 19.0) (n = 39)	0.005
Lowest O₂% (Nadir)	74.3 (25.6) (n = 26)	68.0 (14.5) (n = 42)	0.20
Max ETCO₂ (mm Hg)
Awake	52.0 (6.6) (n = 3)	60.8 (10.5) (n = 8)	0.22
Asleep	43.8 (8.0) (n = 12)	55.9 (8.7) (n = 18)	<0.001
Mean ETCO₂ (mm Hg)
Awake	38.7 (3.0) (n = 2)	39.9 (2.6) (n = 8)	0.57
Asleep	34.0 (4.5) (n = 7)	41.6 (8.1) (n = 5)	0.062
Average ETCO₂ (mm Hg)
REM	35.7 (7.3) (n = 13)	33.0 (7.9) (n = 23)	0.31
Non-REM	34.7 (6.6) (n = 14)	34.3 (8.4) (n = 24)	0.86

AHI, apnea-hypopnea index; RDI, respiratory distress index; OAHI, obstructive apnea-hypopnea index; REM, rapid eye movement sleep phase; non-REM, non-rapid eye movement sleep phase; CAI, central apnea index; O₂, oxygen; ETCO₂, end-tidal carbon dioxide.

* P-values were obtained from unpaired t-test/Wilcoxon-Mann-Whitney test.

ROC curve analysis was done on those continuous variables that were statistically different between groups, as identified by previous univariate analysis. AHI, OAHI REM, OAHI Non-REM, and maximum ETCO2 (asleep) were shown to be “good” predictors for surgical airway management ( Supplemental Figure ). ROC-defined optimal cut points, along with the sensitivity and specificity, were calculated and listed in Table 6 .

Table 6.

ROC Analysis.

Predictors	n	Optimal cut point	Sensitivity at cut point (%)	Specificity at cut point (%)	Area under ROC curve
Max ETCO₂ (asleep)	30	> 49 mm Hg	72	83	0.85
OAHI (REM)	35	> 25.9 events/hr	92	73	0.85
AHI or RDI	70	> 16.9 events/hr	85	72	0.82
OAHI (Non-REM)	35	> 23.6 events/hr	67	91	0.80
Max CO₂*	89	> 57 mm Hg	68	78	0.76
OAHI (Non-Supine REM)	33	> 29.1 events/hr	82	73	0.76
Time O₂ < 90%	62	> 5.6%	62	83	0.72
Max Bicarbonate*	94	> 30 mmol/L	70	62	0.68

CO₂, carbon dioxide; AHI, apnea-hypopnea index; RDI, respiratory distress index; OAHI(REM), obstructive apnea-hypopnea index during rapid eye movement sleep phase; REM, rapid eye movement sleep phase; OAHI (non-REM), obstructive apnea-hypopnea index during non-rapid eye movement sleep phase; O₂, oxygen; ETCO₂, end-tidal carbon dioxide.

*We analyzed all available data for Max CO₂ and Max Bicarbonate while for other variables we analyzed only patients with PSG data available.

Discussion

While the need for surgical airway intervention may be obvious in some patients with RS, optimal treatment for a significant proportion of patients may be less apparent. The decision-making process to cease futile conservative airway interventions and proceed with a surgical airway procedure in these patients lacks consensus and remains controversial. The authors therefore designed the current study with the primary objective of identifying clinical factors in infants with RS who fail a prolonged course of conservative airway management. The current investigation represents the largest single institutional study specifically designed to address this challenging issue of predicting failure of conservative airway management.

The authors’ primary impetus for the current investigation was the growing body of evidence supporting concerns that delayed correction of airway obstruction in the pediatric population may pose a significant increased risk for future problems. Speech and language delays,⁴⁰ hypertension, cardiac dysfunction, and abnormal autonomic system,⁴¹ cor pulmonale, aspiration pneumonia, and/or failure to thrive, hypoxic brain damage, impaired mental development have been documented in young patients with chronic OSA,^16,42 as well as increased risk for hyperactivity and aggression, anxiety and depression,^25,43 and deficits in spatial memory, verbal comprehension, and reasoning,^44–46 although the majority of these studies were done in young children and not in infants. Additionally, care of infants with RS can require prolonged hospitalization, ranging from 10 to 60 days,^47–50 and translate into exorbitant costs of care; moreover, the mortality rates of up to 13.7% for infants with RS are not inconsequential,^{3,16,48,49,51} with incidences as high as 26% for syndromic patients alone.^14,52 With these considerations in mind, we aimed to identify and compare infants with RS managed by our multidisciplinary team after successful conservative airway measures vs. those who failed a prolonged course of conservative treatment and ultimately, underwent a definitive surgical airway procedure. It is important to underscore that operative airway intervention was not performed until a median period of 23.0 days after admission and 14.0 days after PSG. Thus, patients were treated conservatively for a number of weeks before a definitive surgical airway procedure was performed. This provided a unique opportunity to investigate uncharacterized differences between those patients who were successfully managed with conservative measure alone vs. those who ultimately failed and required a definitive surgical airway.

In our study of 122 patients, we identified both categorical and continuous clinical factors associated with failure of conservative airway management. Categorical factors included: (1) greater incidence of systemic anomalies; (2) need for supplemental O2 support. Continuous clinical factors associated with failure of prolonged, conservative airway management included: (1) maximum CO2; and (2) maximum HCO3 on capillary blood gas, as well as; (3) AHI; (4) OAHI during REM; (5) OAHI during non-supine REM sleep; (6) OAHI during non-REM sleep; (7) time with O2sat <90%; and (8) maximum ETCO2 (asleep) during polysomnography ( Table 6 ). While these results seem intuitive, our findings represent the first set of data-supported factors from a study specifically designed to investigate clinical factors that may correlate with failure of conservative therapy, despite weeks of futile efforts.

Other groups have promulgated their indications for surgical airway intervention. While desaturation events are the most frequently reported indication for surgical intervention in RS patients with upper airway obstruction⁵³ our findings suggest this criterion has poor predictive value. Our study did not find a significant difference in the nadir spO2 between groups, with a mean nadir spO2 of 74% in Group 1 and 68.0% in Group 2 (p > 0.05). Similarly, though our study identified that time O2 saturation < 90% to be a significant predictor of failed conservative airway management on univariate analysis (Group 1 vs. Group 2, 1.1% vs. 10.4%, p < 0.005), ROC analysis documented this variable to be only a “fair” predictor (AUC = 0.72).

Markers of carbon dioxide retention have also been described as indicators for surgical airway management in patients with RS.^{2,12,47,51,54–57} Fahradyan and colleagues (2019) reported the use of CBG CO2 in 73 patients with RS and stated that ROC curves with cutoff value of pCO2 level of 49.5 were the best among other variables in terms of specificity and sensitivity combination as an indication for surgical intervention.⁵⁸ Kwan and associates (2021) also noted that a PCO2 cutoff of 55 mmHg (sensitivity = 55.7%; specificity = 91.3%; AUC = 0.78) and a HCO3 cutoff of 30 mEq/L (sensitivity = 56.2%; specificity = 95.0%; AUC = 0.80) were most predictive of patients who “required” a surgical airway.⁵⁹ Neither of these studies specifically documented how the authors determined their surgical airway indications. It is also somewhat disconcerting that most patients (greater than 70%) in these studies “required” a surgical airway, despite the widely held belief that the majority of patients with RS can be successfully managed without a surgical airway. Our own study provides some support for the association of CO2 retention with eventual failure of conservative airway management in patients with RS. Nonetheless, the current study demonstrated that both CBG CO2 and HCO3 were only “fair” and “poor” predictors of prolonged failed conservative airway management, AUC of 0.76 and 0.68, respectively, while maximum ETCO2 while asleep during PSG was one our strongest predictors (AUC = 0.85). Interestingly, the use of maximum ETCO2 as a criterium for surgical intervention in RS is not reported in the literature. In their experience of conservative versus surgically managed RS patients, Runyan et al. found no difference in mean ETCO2 between groups.⁶⁰ Kam and colleagues also found no difference in mean ETCO2 when comparing those managed conservatively vs. surgically.⁶¹ However, neither group independently reported maximum ETCO2, whereas the current study specifically investigated both factors as potential predictors of failed conservative airway management, though mean ETCO2 was not significantly different between groups.

This lack of consensus regarding indications for surgical airway was underscored by a recent survey study by Fan et al., who found a wide range of minimum AHI for surgical intervention, with 78% using a threshold of at least 10 events/hour, and 35% of at least 20 events/hour.⁶² Reddy and associates also provided a literature review revealing mean AHI ranging from 6.4 to 52.6 events/hour in surgical management of RS.⁶³ Runyan and colleagues recommend surgery be considered at an AHI of 20 events/hour based on their 10-year retrospective review,⁶⁰ while Cladis et al. recommend intervention for AHI > 5 events/hour.⁶⁴ Our PSG analysis provides added granularity to the quantification of AHI by independently comparing OAHI during REM and non-REM sleep and when positioned supine and non-supine in REM. Sleep stage impacts tone, with REM sleep associated with hypotonia of the chest wall and upper airway muscles.⁶⁵ Accordingly, OAHI in REM sleep was significantly higher in patients who eventually failed conservative airway management; this factor was also a “good” predictor of the need for surgical intervention by ROC analysis (AUC 0.85, cut point 25.9 events/hour).

The diagnostic appropriateness of PSG in the infant population is not without controversy. Daftary and colleagues recently documented OSA parameters well above the standards set by the American Academy of Sleep Medicine.⁶⁶ Their investigation in healthy full-term infants at a mean age of 19.6 days demonstrated a mean AHI of 14.9 events/hour, though the majority were hypopneic events. While the current study documented AHI cut point of 16.9 events/hour, we also identified OAHI (REM) cut point of 25.9 events/hour and OAHI (non-REM) cut point of 23.6 events/hour. These latter two parameters were not specifically studied by Daftary and associates, but their findings and our own results question the appropriateness of using AHI values of 5 or even 10 events/hour as an indication for surgical airway intervention, as was most used by respondents to the survey performed by Fan and colleagues.⁶¹

The current study has limitations. Due to its retrospective design, our findings are not as robust as those in a prospective RCT. Although all data are gathered prospectively, the authors have not adhered to a strict treatment protocol, as modifications are continuously made in an effort to improve patient outcomes. Other limitations included the relatively small study population and absence of polysomnographic data in all cases. Nonetheless, our sample size of 122 patients is the largest from a single institution to address this issue. In addition, our patient cohort, comprised of roughly 50% syndromic patients, represents a selection bias that must be acknowledged. The authors recognize that many, if not most, patients with nonsyndromic RS succeed with conservative airway management. Nevertheless, the goal of this study was to identify characteristics of patients with RS that ultimately fail conservative airway management, which may guide clinicians in the treatment of those who are not readily or obviously categorized into the conservative or surgical groups. The lack of PSG data in all patients may be viewed as a selection bias, as well. As noted in our Results, PSG data could not be obtained in patients who required prolonged intubation or supplemental O2, which precluded accurate investigations. Thus, the decision to perform a surgical airway procedure on these patients was less controversial. Similarly, those patients who clearly and quickly improved with conservative airway measures did not undergo PSG, which makes intuitive sense. Finally, one must acknowledge the reality in which care for neonates with RS is delivered. Providers are routinely stratifying patients using all available data, some of which have been validated in our analysis. An exhaustive RCT of conservative management vs. early surgical airway intervention for every patient is not realistic, nor is it always what is best for the patient, the family or the utilization of health care resources. Prolonged hospitalizations increase costs and are a source of stress for families. One study reported an average LOS of conservative management of 64 days, higher than our mean LOS for our surgical cohort (54 days) and well above our mean LOS for our conservative group (12 days).³³ Although these authors did not find a significant difference in cost between their surgical and conservative patients, differences between Canadian and American healthcare systems should be recognized. One must remember the purpose of intervention in RS: to reduce mortality and the negative physiologic impact of prolonged airway obstruction. Extended conservative management in the hospital may reduce the chance of a catastrophic event but may not mitigate the physiologic effects of long-term airway obstruction.

Conclusions

Identifying objective clinical parameters that may predict failed conservative airway management in infants with RS remains a challenge, given the realities of clinical medicine. The imperfection of each individual factor is illustrative of the heterogeneity of this patient population. As such, an objective approach that accounts for an array of predictors is more suitable to the disease process. Therefore, for patients who are not easily categorized into a conservative or surgical airway treatment protocol, we propose a set of contributing predictors of failure of conservative management: max ETCO2, AHI, OAHI (REM), OAHI (Non-REM), collectively named, as the acronym, the EARN criteria. These 4 factors individually identify patients with RS who are at greater risk for eventual failure of conservative airway management, despite weeks of these non-invasive measures. Future investigation and validation of these criteria will improve the efficacy and efficiency of managing these challenging patients.

Supplemental Material

sj-docx-1-cpc-10.1177_10556656231224194 - Supplemental material for Predicting Failure of Conservative Airway Management in Infants with Robin Sequence: The EARN Factors

Supplemental material, sj-docx-1-cpc-10.1177_10556656231224194 for Predicting Failure of Conservative Airway Management in Infants with Robin Sequence: The EARN Factors by Jennifer L. McGrath, Esperanza Mantilla-Rivas, Marudeen Aivaz, Monica Manrique, Md Sohel Rana, Hannah R. Crowder, Nathanael S. Oh, Gary F. Rogers and Albert K. Oh in The Cleft Palate Craniofacial Journal

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Esperanza Mantilla-Rivas

Monica Manrique

Hannah R. Crowder

Statement of institutional review board approval and/or statement of conforming to the Declaration of Helsinki

This study was approved by the Institutional Review Board for human research (Pro00014271) at the Children's National Hospital, Washington, DC.

Presentation

Presented, in part, at the 12th Congress of the Asian Pacific Craniofacial Association, December 2018; International Society of Craniofacial Surgery, Virtual Symposium, October 2021; 90th Annual Meeting of the American Society of Plastic Surgeons, October 2021.

Supplemental Material

Supplemental material for this article is available online.

References

Bush

Williams

. Incidence of the Robin Anomalad (Pierre Robin syndrome). Br J Plast Surg. 1983;36(4):434-437. doi:https://doi.org/10.1016/0007-1226(83)90123-6

Evans

Sie

Hopper

Glass

Hing

Cunningham

. Robin sequence: From diagnosis to development of an effective management plan. Pediatrics. 2011;127(5):936-948. doi:https://doi.org/10.1542/peds.2010-2615

Marques

de Sousa

Carneiro

Barbieri

Bettiol

Pereira Gutierrez

. Clinical experience with infants with Robin sequence: A prospective study. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2001;38(2):171-178. doi:https://doi.org/10.1597/1545-1569_2001_038_0171_CEWIWR_2.0.CO_2

Anderson

Cole

Chuo

Slator

. Home management of upper airway obstruction in Pierre Robin sequence using a nasopharyngeal airway. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2007;44(3):269-273. doi:https://doi.org/10.1597/06-020

Marques

De Barros Almeida Peres

Bettiol

Barbieri

Andrea

De Souza

. Growth of children with isolated Robin sequence treated by nasopharyngeal intubation: Importance of a hypercaloric diet. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2004;41(1):53-58. doi:https://doi.org/10.1597/02-043

Mondini

CCSD

Marques

Fontes

CMB

Thomé

. Nasopharyngeal intubation in Robin sequence: Technique and management. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2009;46(3):258-261. doi:https://doi.org/10.1597/08-042.1

Delorme

Larocque

Caouette-Laberge

. Innovative surgical approach for the Pierre Robin anomalad: Subperiosteal release of the floor of the mouth musculature. Plast Reconstr Surg. 1989;83(6):960-966.

Abramowicz

Bacic

Mulliken

Rogers

. Validation of the GILLS score for tongue-lip adhesion in robin sequence patients. Journal of Craniofacial Surgery. 2012;23(2):382-386. doi:https://doi.org/10.1097/SCS.0b013e318240fc7b

Argamaso

. Glossopexy for upper airway obstruction in Robin sequence. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 1992;29(3):232-238. doi:https://doi.org/10.1597/1545-1569_1992_029_0232_GFUAOI_2.3.CO_2

10.

Cozzi

Totonelli

Frediani

Zani

Spagnol

Cozzi

. The effect of glossopexy on weight velocity in infants with Pierre Robin syndrome. J Pediatr Surg. 2008;43(2):296-298. doi:https://doi.org/10.1016/J.JPEDSURG.2007.10.015

11.

Denny

Amm

Schaefer

. Outcomes of tongue-lip adhesion for neonatal respiratory distress caused by Pierre Robin sequence. J Craniofac Surg. 2004;15(5):541. doi:https://doi.org/10.1097/00001665-200409000-00023

12.

Hoffman

. Outcome of tongue-lip plication in patients with severe Pierre Robin sequence. J Craniofac Surg. 2003;14(5):602-608. doi:https://doi.org/10.1097/00001665-200309000-00002.

13.

Kirschner

Low

Randall

, et al.

Surgical airway management in Pierre Robin sequence: Is there a role for tongue-lip adhesion?

The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2003;40(1):13-18. doi:https://doi.org/10.1597/1545-1569_2003_040_0013_SAMIPR_2.0.CO_2

14.

Parsons

Smith

. A modified tongue-lip adhesion for Pierre Robin anomalad. Cleft Palate J. 1980;17(2):144-147.

15.

Demke

Bassim

Patel

, et al. Parental perceptions and morbidity: Tracheostomy and Pierre Robin sequence. Int J Pediatr Otorhinolaryngol. 2008;72(10):1509-1516. doi:https://doi.org/10.1016/J.IJPORL.2008.07.002.

16.

Tomaski

Zalzal

Saal

. Airway obstruction in the Pierre Robin sequence. Laryngoscope. 1995;105(2):111-114. doi:https://doi.org/10.1288/00005537-199502000-00001

17.

Burstein

Williams

. Mandibular distraction osteogenesis in Pierre Robin sequence: Application of a new internal single-stage resorbable device. Plast Reconstr Surg. 2005;115(1):61-69.

18.

Dauria

Marsh

. Mandibular distraction osteogenesis for Pierre Robin sequence: What percentage of neonates need it? J Craniofac Surg. 2008;19(5):1237-1243. doi:https://doi.org/10.1097/SCS.0B013E3181843293

19.

Denny

Kalantarian

. Mandibular distraction in neonates: A strategy to avoid tracheostomy. Plast Reconstr Surg. 2002;109(3):896-904. doi:https://doi.org/10.1097/00006534-200203000-00011

20.

Denny

. Distraction osteogenesis in Pierre Robin neonates with airway obstruction. Clin Plast Surg. 2004;31(2):221-229. doi:https://doi.org/10.1016/S0094-1298(03)00131-7

21.

Denny

Talisman

Hanson

Recinos

. Mandibular distraction osteogenesis in very young patients to correct airway obstruction. Plast Reconstr Surg. 2001;108(2):302-311. doi:https://doi.org/10.1097/00006534-200108000-00004

22.

Monasterio

Drucker

Molina

Ysunza

. Distraction osteogenesis in Pierre Robin sequence and related respiratory problems in children. J Craniofac Surg. 2002;13(1):79-83. doi:https://doi.org/10.1097/00001665-200201000-00018.

23.

Monasterio

Molina

Berlanga

, et al. Swallowing disorders in Pierre Robin sequence: Its correction by distraction. Journal of Craniofacial Surgery. 2004;15(6):934-941. doi:https://doi.org/10.1097/00001665-200411000-00009.

24.

Wittenborn

Panchal

Marsh

Sekar

Gurley

. Neonatal distraction surgery for micrognathia reduces obstructive apnea and the need for tracheotomy. J Craniofac Surg. 2004;15(4):623-630. doi:https://doi.org/10.1097/00001665-200407000-00018.

25.

Chervin

Archbold

Dillon

, et al. Inattention, hyperactivity, and symptoms of sleep-disordered breathing. Pediatrics. 2002;109(3):449-456. doi:https://doi.org/10.1542/PEDS.109.3.449

26.

Rosen

Storfer-Isser

Gerry Taylor

Lester Kirchner

Emancipator

Redline

. Increased behavioral morbidity in school-aged children with sleep-disordered breathing. Pediatrics. 2004;114(6):1640-1648. doi:https://doi.org/10.1542/PEDS.2004-0103.

27.

O’Brien

Mervis

Holbrook

, et al. Neurobehavioral implications of habitual snoring in children. Pediatrics. 2004;114(1):44-49. doi:https://doi.org/10.1542/PEDS.114.1.44

28.

Gottlieb

Chase

Vezina

, et al. Sleep-disordered breathing symptoms are associated with poorer cognitive function in 5-year-old children. J Pediatr. 2004;145(4):458-464. doi:https://doi.org/10.1016/J.JPEDS.2004.05.039.

29.

Venekamp

Hearne

Chandrasekharan

Blackshaw

Lim

Schilder

AGM

. Tonsillectomy or adenotonsillectomy versus non-surgical management for obstructive sleep-disordered breathing in children. Cochrane Database Syst Rev. 2015;2015(10). doi:https://doi.org/10.1002/14651858.CD011165.PUB2

30.

Amin

Carroll

Jeffries

, et al. Twenty-four-hour ambulatory blood pressure in children with sleep-disordered breathing. Am J Respir Crit Care Med. 2004;169(8):950-956. doi:https://doi.org/10.1164/RCCM.200309-1305OC

31.

Amin

Kimball

Kalra

, et al. Left ventricular function in children with sleep-disordered breathing. Am J Cardiol. 2005;95(6):801-804. doi:https://doi.org/10.1016/J.AMJCARD.2004.11.044

32.

Liao

Rodriguez-Colon

, et al. Sleep-disordered breathing and cardiac autonomic modulation in children. Sleep Med. 2010;11(5):484-488. doi:https://doi.org/10.1016/J.SLEEP.2009.11.012

33.

Lee

Forrest

. Pierre Robin sequence: Cost-analysis and qualitative assessment of 89 patients at the hospital for sick children. Plastic Surgery. 2019;27(1):14. doi:https://doi.org/10.1177/2292550318767922

34.

Han

Seruya

Zalzal

Preciado

. Natural” decannulation in patients with Robin sequence and severe airway obstruction. Ann Otol Rhinol Laryngol. 2012;121(1):44-50. doi:https://doi.org/10.1177/000348941212100108

35.

Albino

Wood

Han

, et al. Clinical factors associated with the non-operative airway management of patients with Robin sequence. Arch Plast Surg. 2016;43(6):506-511. doi:https://doi.org/10.5999/APS.2016.43.6.506

36.

Nino

Aziz

Weiss

, et al. Defining Age-related OSA Features in Robin Sequence Using Polysomnographic-based Analyses of Respiratory Arousal Responses and Gas-exchange Parameters. https://doi.org/101177/10556656211055017. 2021;60(2):142-150. doi:10.1177/10556656211055017.

37.

Rogers

Murthy

LaBrie

Mulliken

. The GILLS score: Part I. Patient selection for tongue-lip adhesion in Robin sequence. Plast Reconstr Surg. 2011;128(1):243-251. doi:https://doi.org/10.1097/PRS.0B013E318217420D

38.

. Assessing the accuracy of diagnostic tests. Shanghai Arch Psychiatry. 2018;30(3):207. doi:https://doi.org/10.11919/J.ISSN.1002-0829.218052

39.

Youden

. Index for rating diagnostic tests. Cancer. 1950;3(1):32-35. doi:https://doi.org/10.1002/1097-0142(1950)3:1<32::aid-cncr2820030106>3.0.co;2-3.

40.

Mohammed

Park

Bogaardt

Docking

. The impact of childhood obstructive sleep apnea on speech and oral language development: A systematic review. Sleep Med. 2021;81:144-153. doi:https://doi.org/10.1016/J.SLEEP.2021.02.015

41.

Muzumdar

Arens

. Physiological effects of obstructive sleep apnea syndrome in childhood. Respir Physiol Neurobiol. 2013;188(3):370-382. doi:https://doi.org/10.1016/J.RESP.2013.05.006

42.

Kapp-Simon

Krueckeberg

. Mental development in infants with cleft lip and/or palate. The Cleft Palate-Craniofacial Journal: Official Publication of the American Cleft Palate-Craniofacial Association. 2000;37(1):65-70. doi:https://doi.org/10.1597/1545-1569_2000_037_0065_MDIIWC_2.3.CO_2

43.

Stein

Mendelsohn

Obermeyer

Amromin

Benca

. Sleep and behavior problems in school-aged children. Pediatrics. 2001;107(4):e60. doi:https://doi.org/10.1542/PEDS.107.4.E60

44.

Spruyt

Capdevila

Kheirandish-Gozal

Gozal

. Inefficient or insufficient encoding as potential primary deficit in neurodevelopmental performance among children with OSA. Dev Neuropsychol. 2009;34(5):601. doi:https://doi.org/10.1080/87565640903133566

45.

Kheirandish-Gozal

De Jong

Spruyt

Chamuleau

SAJ

Gozal

. Obstructive sleep apnoea is associated with impaired pictorial memory task acquisition and retention in children. Eur Respir J. 2010;36(1):164-169. doi:https://doi.org/10.1183/09031936.00114209

46.

Chandrakantan

Adler

. Pediatric obstructive sleep apnea and neurocognition. Anesthesiol Clin. 2020;38(3):693-707. doi:https://doi.org/10.1016/J.ANCLIN.2020.05.004.

47.

Wagener

Rayatt

Tatman

Gornall

Slator

. Management of infants with Pierre Robin sequence. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2003;40(2):180-185. doi:https://doi.org/10.1597/1545-1569_2003_040_0180_moiwpr_2.0.co_2

48.

Bull

Givan

Sadove

Bixler

Hearn

. Improved outcome in Pierre Robin sequence: Effect of multidisciplinary evaluation and management. Pediatrics. 1990;86(2):294-301.

49.

Cruz

Kerschner

Beste

Conley

. Pierre Robin sequences: Secondary respiratory difficulties and intrinsic feeding abnormalities. Laryngoscope. 1999;109(10):1632-1636. doi:https://doi.org/10.1097/00005537-199910000-00016

50.

Matsas

Thomson

Goodacre

. Management of infants with Pierre Robin sequence. The Cleft Palate-Craniofacial Journal : Official Publication of the American Cleft Palate-Craniofacial Association. 2004;41(2):219. doi:https://doi.org/10.1597/1545-1569_2003_040_0180_MOIWPR_2.0.CO_2

51.

Caouette-Laberge

Bayet

Larocque

. The Pierre Robin sequence: Review of 125 cases and evolution of treatment modalities. Plast Reconstr Surg. 1994;93(5):934-942.

52.

Hsieh

Woo

. Pierre Robin sequence. Clin Plast Surg. 2019;46(2):249-259. doi:https://doi.org/10.1016/J.CPS.2018.11.010

53.

Paes

van Nunen

DPF

Speleman

, et al. A pragmatic approach to infants with Robin sequence: A retrospective cohort study and presence of a treatment algorithm. Clin Oral Investig. 2015;19(8):2101-2114. doi:https://doi.org/10.1007/S00784-015-1407-6

54.

Augarten

Sagy

Yahav

Barzilay

. Management of upper airway obstruction in the Pierre Robin syndrome. Br J Oral Maxillofac Surg. 1990;28(2):105-108. doi:https://doi.org/10.1016/0266-4356(90)90133-6

55.

de Buys Roessingh

Herzog

Hohlfeld

. Respiratory distress in Pierre Robin: Successful use of pharyngeal tube. J Pediatr Surg. 2007;42(9):1495-1499. doi:https://doi.org/10.1016/J.JPEDSURG.2007.04.024.

56.

Freed

Pearlman

Brown

Barot

. Polysomnographic indications for surgical intervention in Pierre Robin sequence: Acute airway management and follow-up studies after repair and take-down of tongue-lip adhesion. Cleft Palate J. 1988;25(2):151-155.

57.

Kochel

Meyer-Marcotty

Wirbelauer

, et al. Treatment modalities of infants with upper airway obstruction–review of the literature and presentation of novel orthopedic appliances. The Cleft Palate-Craniofacial Journal: Official Publication of the American Cleft Palate-Craniofacial Association. 2011;48(1):44-55. doi:https://doi.org/10.1597/08-273

58.

Fahradyan

Azadgoli

Tsuha

Urata

Francis

. A single lab test to aid Pierre Robin sequence severity diagnosis. The Cleft Palate-Craniofacial Journal: Official Publication of the American Cleft Palate-Craniofacial Association. 2019;56(3):298-306. doi:https://doi.org/10.1177/1055665618778400

59.

Kwan

Ebert

Roby

Scott

. Detection of chronic hypoventilation among infants with Robin sequence using capillary blood gas sampling. Laryngoscope. 2021;131(12):2789-2794. doi:https://doi.org/10.1002/LARY.29594

60.

Runyan

Uribe-Rivera

Tork

, et al. Management of Airway Obstruction in Infants With Pierre Robin Sequence. Plastic and Reconstructive Surgery Global Open. 2018;6(5):e1688. doi:https://doi.org/10.1097/GOX.0000000000001688

61.

Kam

McKay

MacLean

Witmans

Spier

Mitchell

. Surgical versus nonsurgical interventions to relieve upper airway obstruction in children with Pierre Robin sequence. Can Respir J. 2015;22(3):171-175. doi:https://doi.org/10.1155/2015/798076.

62.

Fan

Mandelbaum

Buro

, et al. Current Trends in Surgical Airway Management of Neonates with Robin Sequence. Plastic and Reconstructive Surgery Global Open. 2018;6(11):e1973. doi:https://doi.org/10.1097/GOX.0000000000001973

63.

Reddy

. Evaluation of upper airway obstruction in infants with Pierre Robin sequence and the role of polysomnography - review of current evidence. Paediatr Respir Rev. 2016;17:80-87. doi:https://doi.org/10.1016/j.prrv.2015.10.001

64.

Cladis

Kumar

Grunwaldt

Otteson

Ford

Losee

. Pierre Robin sequence: A perioperative review. Anesth Analg. 2014;119(2):400-412. doi:https://doi.org/10.1213/ANE.0000000000000301

65.

Katz

Mitchell

. D’Ambrosio CM. Obstructive Sleep Apnea in Infants. https://doi.org/101164/rccm201108-1455CI. 2012;185(8):805-816. doi:10.1164/RCCM.201108-1455CI.

66.

Daftary

Jalou

Shively

Slaven

Davis

. Polysomnography reference values in healthy newborns. J Clin Sleep Med. 2019;15(3):437-443.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.42 MB